Image model comparison · 2 models · Updated Nov 2, 2026

GPT Image 2.5 Flare vs Arrow 2

Newest: Arrow 2 · Self-hosting: No open weights · Most API hosts: GPT Image 2.5 Flare. Not enough arena votes yet for a quality score, so this is a spec comparison.

OpenAI
GPT Image 2.5 Flare
Released Sep 8, 2026New
- Text to image
- Image editing
- Arena1,482
- From—
- Max size—
Full specs Remove GPT Image 2.5 Flare
QuiverAI
Arrow 2
Released Sep 16, 2026New
- Text to image
- Image editing
- Arena—
- From—
- Max size—
Full specs Remove Arrow 2
Add a model
Make it a three-way comparison.
Third model

01 — At a glance

Where they differ

Quality is the arena score from blind, side-by-side human votes. Price, length and resolution come from each lab’s own listings.

NewestArrow 2GPT Image 2.5 Flare Sep 8, 2026 · Arrow 2 Sep 16, 2026
Widest inputsSame inputsGPT Image 2.5 Flare: Text prompts, Reference images · Arrow 2: Text prompts, Reference images
Image editingBoth edit imagesTakes a reference image as input: GPT Image 2.5 Flare yes · Arrow 2 yes
Self-hostingNo open weightsBoth are available only through APIs and apps
Most API hostsGPT Image 2.5 FlareGPT Image 2.5 Flare: 4 · Arrow 2: 1

02 — Side by side

Every spec in one table

Dashes mean the data is not published.

GPT Image 2.5 Flare vs Arrow 2 specifications side by side
Specification	GPT Image 2.5 FlareOpenAI	Arrow 2QuiverAI
Quality (blind human votes)
Text to image arenaElo-style score · higher is better	1,401 ±5#2 of 82 · 39,218 votes	—
Image editing arenaElo-style score · higher is better	1,482 ±5#2 of 61 · 87,941 votes	—
Price
How it is billed	Billed per token: $30 per million output image tokens.	—
Output
Produces	Images	Images
Images per request	Up to 10	—
Inputs
Text prompts	Yes	Yes
Reference images (editing)	Yes	Yes
Max reference images	16	—
Access
Weights	Proprietary	Proprietary
API hosts	4 (best)	1
Released	Sep 8, 2026	Sep 16, 2026

03 — Questions

Which should you choose?

Which is better, GPT Image 2.5 Flare or Arrow 2?

Neither model has enough arena votes for a reliable quality score yet, so this page compares specs only. Run the same three or four prompts through each and judge the results.

What is the difference between GPT Image 2.5 Flare and Arrow 2?

GPT Image 2.5 Flare is an image model from OpenAI (Sep 8, 2026), taking text prompts and reference images and producing images. Arrow 2 is an image model from QuiverAI (Sep 16, 2026), taking text prompts and reference images and producing images.

Which is newer?

GPT Image 2.5 Flare Sep 8, 2026; Arrow 2 Sep 16, 2026.

Which can edit an existing image?

Takes a reference image as input: GPT Image 2.5 Flare yes; Arrow 2 yes.

Can I run any of these on my own hardware?

No open weights. Both are available only through APIs and apps.